merge-vads.cc File Reference

#include "base/kaldi-common.h"
#include "util/common-utils.h"
#include "matrix/kaldi-matrix.h"
#include "util/stl-utils.h"

Include dependency graph for merge-vads.cc:

Go to the source code of this file.

Namespaces
	kaldi
	This code computes Goodness of Pronunciation (GOP) and extracts phone-level pronunciation feature for mispronunciations detection tasks, the reference:

Functions
void	PrepareMap (const std::string map_rxfilename, unordered_map< std::pair< int32, int32 >, int32, PairHasher< int32 > > *map)
	PrepareMap creates a mapping between the pairs of VAD decisions and the output label. More...
int	main (int argc, char *argv[])

Function Documentation

main()

int main	(	int	argc,
		char *	argv[]
	)

Definition at line 85 of file merge-vads.cc.

References ParseOptions::GetArg(), rnnlm::i, KALDI_ERR, KALDI_LOG, KALDI_WARN, ParseOptions::NumArgs(), kaldi::PrepareMap(), ParseOptions::PrintUsage(), ParseOptions::Read(), and ParseOptions::Register().

                                 {
  using namespace kaldi;
  typedef kaldi::int32 int32;
  try {
    const char *usage =
      "This program merges two archives of per-frame weights representing\n"
      "voice activity decisions.  By default, the program assumes that the\n"
      "input vectors consist of floats that are 0.0 if a frame is judged\n"
      "as nonspeech and 1.0 if it is considered speech.  The default\n"
      "behavior produces a frame-level decision of 1.0 if both input frames\n"
      "are 1.0, and 0.0 otherwise.  Additional classes (e.g., 2.0 for music)\n"
      "can be handled using the \"map\" option.\n"
      "\n"
      "Usage: merge-vads [options] <vad-rspecifier-1> <vad-rspecifier-2>\n"
      "    <vad-wspecifier>\n"
      "e.g.: merge-vads [options] scp:vad_energy.scp scp:vad_gmm.scp\n"
      "    ark:vad.ark\n"
      "See also: compute-vad-from-frame-likes, compute-vad, ali-to-post,\n"
      "post-to-weights\n";

    ParseOptions po(usage);
    std::string map_rxfilename;
    po.Register("map", &map_rxfilename, "This table specifies a mapping "
      "between the labels of the frame-level decisions in the first and "
      "second input archives to the integer output label.");

    po.Read(argc, argv);
    if (po.NumArgs() != 3) {
      po.PrintUsage();
      exit(1);
    }

    unordered_map<std::pair<int32, int32>, int32, PairHasher<int32> > map;
    PrepareMap(map_rxfilename, &map);
    SequentialBaseFloatVectorReader first_vad_reader(po.GetArg(1));
    RandomAccessBaseFloatVectorReader second_vad_reader(po.GetArg(2));
    BaseFloatVectorWriter vad_writer(po.GetArg(3));

    int32 num_done = 0, num_err = 0;
    for (;!first_vad_reader.Done(); first_vad_reader.Next()) {
      std::string utt = first_vad_reader.Key();
      Vector<BaseFloat> vad1(first_vad_reader.Value());
      if (!second_vad_reader.HasKey(utt)) {
        KALDI_WARN << "No vector for utterance " << utt;
        num_err++;
        continue;
      }
      Vector<BaseFloat> vad2(second_vad_reader.Value(utt));
      if (vad1.Dim() != vad2.Dim()) {
        KALDI_WARN << "VAD length mismatch for utterance " << utt;
        num_err++;
        continue;
      }
      Vector<BaseFloat> vad_result(vad1.Dim());
      for (int32 i = 0; i < vad1.Dim(); i++) {
        std::pair<int32, int32> key(static_cast<int32>(vad1(i)),
          static_cast<int32>(vad2(i)));
        unordered_map<std::pair<int32, int32>, int32,
          PairHasher<int32> >::const_iterator iter = map.find(key);
        if (iter == map.end()) {
          KALDI_ERR << "Map is missing combination "
                    << vad1(i) << " and " << vad2(i);
        } else {
          vad_result(i) = iter->second;
        }
      }

      vad_writer.Write(utt, vad_result);
      num_done++;
    }
    KALDI_LOG << "Merged voice activity detection decisions; "
              << "processed " << num_done << " utterances successfully; "
              << num_err << " had errors.";
    return (num_done != 0 ? 0 : 1);
  } catch(const std::exception &e) {
    std::cerr << e.what();
    return -1;
  }
}